But, if you go lower impact loads, will the results translate up the scale? In other words, how can you predict the linearity? Some designs would, I imagine, not react in a linear fashion. Perhaps a set of three hits, at different lower levels, just shy of the 'damage stuff' point, to try and get some points that could be extrapolated from?
[/b]
An excellent point, sir.

Testing has shown that head loads are not a linear function of impact, but it's close enough that proportionality is used frequently as a rule of thumb for estimating absolute levels. For comparative purposes it's a non-issue, i.e. if a product has a certain head load reduction at 50Gs, say, it's reasonable to see the same general level of performance at different impacts.

There is plenty of precedent. At both the 2002 and 2004 SAE conferences Melvin, Begeman and Foster authored papers comparing the performance of various H&N restraints on the 50G sled at WSU. Because the issue was relative performance, there was no question of absolute load levels at different impacts.

You would be doing exactly the same thing here, with the same sled and the same dummy. The only difference is that you rotate everything 90 degrees and dial down the impact.